Get the total number of bytes used for each sender in an Mbox email file

mbox-size.rb

#!/usr/bin/env ruby

emails = {}
email_length = 0
last_email_from = nil

File.open(ARGV[0], "r:ASCII-8BIT").each_line do |line|
  if line.start_with?('From ')
    unless last_email_from.nil?
      emails[last_email_from] ||= 0
      emails[last_email_from] += email_length
      last_email_from = nil
    end
    email_length = 0
  end
  if line.start_with?('From: ')
    parsed_address = line.sub(/^From: /, '')
    if parsed_address =~ /<.+>/
      parsed_address = parsed_address.split('<')[1].split('>').first
    end
    last_email_from = parsed_address
  end
  email_length += line.length
end

emails.sort_by{|from, length| length}.reverse.each do |result|
  puts "#{result.last} #{result.first}"
end